Statistical treatment of grain-size curves and empirical distributions: densities as compositions?

The preceding two editions of CoDaWork included talks on the possible considerationof densities as infinite compositions: Egozcue and D´ıaz-Barrero (2003) extended theEuclidean structure of the simplex to a Hilbert space structure of the set of densitieswithin a bounded interval, and van den Boogaart (2005) generalized this to the setof densities bounded by an arbitrary reference density. From the many variations ofthe Hilbert structures available, we work with three cases. For bounded variables, abasis derived from Legendre polynomials is used. For variables with a lower bound, westandardize them with respect to an exponential distribution and express their densitiesas coordinates in a basis derived from Laguerre polynomials. Finally, for unboundedvariables, a normal distribution is used as reference, and coordinates are obtained withrespect to a Hermite-polynomials-based basis.To get the coordinates, several approaches can be considered. A numerical accuracyproblem occurs if one estimates the coordinates directly by using discretized scalarproducts. Thus we propose to use a weighted linear regression approach, where all k-order polynomials are used as predictand variables and weights are proportional to thereference density. Finally, for the case of 2-order Hermite polinomials (normal reference)and 1-order Laguerre polinomials (exponential), one can also derive the coordinatesfrom their relationships to the classical mean and variance.Apart of these theoretical issues, this contribution focuses on the application of thistheory to two main problems in sedimentary geology: the comparison of several grainsize distributions, and the comparison among different rocks of the empirical distribution of a property measured on a batch of individual grains from the same rock orsediment, like their composition

Distributions on the simplex

The simplex, the sample space of compositional data, can be structured as a real Euclidean space. This fact allows to work with the coefficients with respect to an orthonormal basis. Over these coefficients we apply standard real analysis, inparticular, we define two different laws of probability trought the density function and we study their main properties

Comparing correspondence analysis and the log-ratio alternative for representing categorical data

We compare correspondance análisis to the logratio approach based on compositional data. We also compare correspondance análisis and an alternative approach using Hellinger distance, for representing categorical data in a contingency table. We propose a coefficient which globally measures the similarity between these approaches. This coefficient can be decomposed into several components, one component for each principal dimension, indicating the contribution of the dimensions to the difference between the two representations. These three methods of representation can produce quite similar results. One illustrative example is given

Meaning of the λ parameter of skew–normal and log–skew normal distributions in fluid geochemistry

The literature related to skew–normal distributions has grown rapidly in recent yearsbut at the moment few applications concern the description of natural phenomena withthis type of probability models, as well as the interpretation of their parameters. Theskew–normal distributions family represents an extension of the normal family to whicha parameter (λ) has been added to regulate the skewness. The development of this theoreticalfield has followed the general tendency in Statistics towards more flexible methodsto represent features of the data, as adequately as possible, and to reduce unrealisticassumptions as the normality that underlies most methods of univariate and multivariateanalysis. In this paper an investigation on the shape of the frequency distribution of thelogratio ln(Cl−/Na+) whose components are related to waters composition for 26 wells,has been performed. Samples have been collected around the active center of Vulcanoisland (Aeolian archipelago, southern Italy) from 1977 up to now at time intervals ofabout six months. Data of the logratio have been tentatively modeled by evaluating theperformance of the skew–normal model for each well. Values of the λ parameter havebeen compared by considering temperature and spatial position of the sampling points.Preliminary results indicate that changes in λ values can be related to the nature ofenvironmental processes affecting the data

A Log-linear model of grain size influence on the geochemistry of sediments

Sediment composition is mainly controlled by the nature of the source rock(s), and chemical (weathering) and physical processes (mechanical crushing, abrasion, hydrodynamic sorting) during alteration and transport. Although the factors controlling these processes are conceptually well understood, detailed quantification of compositional changes induced by a single process are rare, as are examples where the effects of several processes can be distinguished. The present study was designed to characterize the role of mechanical crushing and sorting in the absence of chemical weathering. Twenty sediment samples were taken from Alpine glaciers that erode almost pure granitoid lithologies. For each sample, 11 grain-size fractions from granules to clay (ø grades &-1 to &9) were separated, and each fraction was analysed for its chemical composition.The presence of clear steps in the box-plots of all parts (in adequate ilr and clr scales) against ø is assumed to be explained by typical crystal size ranges for the relevant mineral phases. These scatter plots and the biplot suggest a splitting of the full grain size range into three groups: coarser than ø=4 (comparatively rich in SiO2, Na2O, K2O, Al2O3, and dominated by “felsic” minerals like quartz and feldspar), finer than ø=8 (comparatively rich in TiO2, MnO, MgO, Fe2O3, mostly related to “mafic” sheet silicates like biotite and chlorite), and intermediate grains sizes (4≤ø &8; comparatively rich in P2O5 and CaO, related to apatite, some feldspar).To further test the absence of chemical weathering, the observed compositions were regressed against three explanatory variables: a trend on grain size in ø scale, a step function for ø≥4, and another for ø≥8. The original hypothesis was that the trend could be identified with weathering effects, whereas each step function would highlight those minerals with biggest characteristic size at its lower end. Results suggest that this assumption is reasonable for the step function, but that besides weathering some other factors (different mechanical behavior of minerals) have also an important contribution to the trend.Key words: sediment, geochemistry, grain size, regression, step function

Aplicación de técnicas de minería de datos a los log generados por CATEDU

Extracción de conocimiento de los log generados por un servidor web aplicando técnicas de minería de datos.

Suitability, success and sinks: how do predictions of nesting distributions relate to fitness parameters in high arctic waders?

AimAlthough habitat suitability maps derived from species distribution models (SDMs) are often assumed to highlight locations that can sustain healthy populations over time, the relationship between suitability scores and fitness parameters has rarely been tested thoroughly. LocationZackenberg Valley, north-east Greenland. MethodsUsing 14years of data (1997-2010) representing three wader species (dunlin Calidris alpina, sanderling Calidris alba and ruddy turnstone Arenaria interpres), we tested the relationships between modelled suitability and fitness parameters at nesting locations. ResultsAmong the three species examined, only the ruddy turnstone exhibited significant relationships between suitability and nest success, but over time rather than space. During years with extensive snow cover in the landscape, the nesting sites of ruddy turnstone occurred in different habitats than were typically used across years. Moreover, in years with extensive snow cover, the ruddy turnstone initiated nests later and suffered from higher egg predation rates. Main conclusionOur results suggest that SDMs derived from species occurrences that include years of low reproductive success may over-estimate the potential suitable habitat in the landscape. Whenever possible, variation in reproductive success should be considered when building models to inform species' response to environmental change. species' response to environmental change.

Monitoring procedures in environmental geochemistry and compositional data analysis theory

First discussion on compositional data analysis is attributable to Karl Pearson, in 1897. However, notwithstanding the recent developments on algebraic structure of the simplex, more than twenty years after Aitchison’s idea of log-transformations of closed data, scientific literature is again full of statistical treatments of this type of data by using traditional methodologies. This is particularly true in environmental geochemistry where besides the problem of the closure, the spatial structure (dependence) of the data have to be considered. In this work we propose the use of log-contrast values, obtained by asimplicial principal component analysis, as LQGLFDWRUV of given environmental conditions. The investigation of the log-constrast frequency distributions allows pointing out the statistical laws able togenerate the values and to govern their variability. The changes, if compared, for example, with the mean values of the random variables assumed as models, or other reference parameters, allow definingmonitors to be used to assess the extent of possible environmental contamination. Case study on running and ground waters from Chiavenna Valley (Northern Italy) by using Na+, K+, Ca2+, Mg2+, HCO3-, SO4 2- and Cl- concentrations will be illustrated

Species distributions and interactions in a changing world : from understanding to forecasting

Les écosystèmes fournissent de nombreuses ressources et services écologiques qui sont utiles à la population humaine. La biodiversité est une composante essentielle des écosystèmes et maintient de nombreux services. Afin d'assurer la permanence des services écosystémiques, des mesures doivent être prises pour conserver la biodiversité. Dans ce but, l'acquisition d'informations détaillées sur la distribution de la biodiversité dans l'espace est essentielle. Les modèles de distribution d'espèces (SDMs) sont des modèles empiriques qui mettent en lien des observations de terrain (présences ou absences d'une espèce) avec des descripteurs de l'environnement, selon des courbes de réponses statistiques qui décrive la niche réalisée des espèces. Ces modèles fournissent des projections spatiales indiquant les lieux les plus favorables pour les espèces considérées. Le principal objectif de cette thèse est de fournir des projections plus réalistes de la distribution des espèces et des communautés en montagne pour le climat présent et futur en considérant non-seulement des variables abiotiques mais aussi biotiques. Les régions de montagne et l'écosystème alpin sont très sensibles aux changements globaux et en même temps assurent de nombreux services écosystémiques. Cette thèse est séparée en trois parties : (i) fournir une meilleure compréhension du rôle des interactions biotiques dans la distribution des espèces et l'assemblage des communautés en montagne (ouest des Alpes Suisses), (ii) permettre le développement d'une nouvelle approche pour modéliser la distribution spatiale de la biodiversité, (iii) fournir des projections plus réalistes de la distribution future des espèces ainsi que de la composition des communautés. En me focalisant sur les papillons, bourdons et plantes vasculaires, j'ai détecté des interactions biotiques importantes qui lient les espèces entre elles. J'ai également identifié la signature du filtre de l'environnement sur les communautés en haute altitude confirmant l'utilité des SDMs pour reproduire ce type de processus. A partir de ces études, j'ai contribué à l'amélioration méthodologique des SDMs dans le but de prédire les communautés en incluant les interactions biotiques et également les processus non-déterministes par une approche probabiliste. Cette approche permet de prédire non-seulement la distribution d'espèces individuelles, mais également celle de communautés dans leur entier en empilant les projections (S-SDMs). Finalement, j'ai utilisé cet outil pour prédire la distribution d'espèces et de communautés dans le passé et le futur. En particulier, j'ai modélisé la migration post-glaciaire de Trollius europaeus qui est à l'origine de la structure génétique intra-spécifique chez cette espèce et évalué les risques de perte face au changement climatique. Finalement, j'ai simulé la distribution des communautés de bourdons pour le 21e siècle afin d'évaluer les changements probables dans ce groupe important de pollinisateurs. La diversité fonctionnelle des bourdons va être altérée par la perte d'espèces spécialistes de haute altitude et ceci va influencer la pollinisation des plantes en haute altitude. - Ecosystems provide a multitude of resources and ecological services, which are useful to human. Biodiversity is an essential component of those ecosystems and guarantee many services. To assure the permanence of ecosystem services for future generation, measure should be applied to conserve biodiversity. For this purpose, the acquisition of detailed information on how biodiversity implicated in ecosystem function is distributed in space is essential. Species distribution models (SDMs) are empirical models relating field observations to environmental predictors based on statistically-derived response surfaces that fit the realized niche. These models result in spatial predictions indicating locations of the most suitable environment for the species and may potentially be applied to predict composition of communities and their functional properties. The main objective of this thesis was to provide more accurate projections of species and communities distribution under current and future climate in mountains by considering not solely abiotic but also biotic drivers of species distribution. Mountain areas and alpine ecosystems are considered as particularly sensitive to global changes and are also sources of essential ecosystem services. This thesis had three main goals: (i) a better ecological understanding of biotic interactions and how they shape the distribution of species and communities, (ii) the development of a novel approach to the spatial modeling of biodiversity, that can account for biotic interactions, and (iii) ecologically more realistic projections of future species distributions, of future composition and structure of communities. Focusing on butterfly and bumblebees in interaction with the vegetation, I detected important biotic interactions for species distribution and community composition of both plant and insects along environmental gradients. I identified the signature of environmental filtering processes at high elevation confirming the suitability of SDMs for reproducing patterns of filtering. Using those case-studies, I improved SDMs by incorporating biotic interaction and accounting for non-deterministic processes and uncertainty using a probabilistic based approach. I used improved modeling to forecast the distribution of species through the past and future climate changes. SDMs hindcasting allowed a better understanding of the spatial range dynamic of Trollius europaeus in Europe at the origin of the species intra-specific genetic diversity and identified the risk of loss of this genetic diversity caused by climate change. By simulating the future distribution of all bumblebee species in the western Swiss Alps under nine climate change scenarios for the 21st century, I found that the functional diversity of this pollinator guild will be largely affected by climate change through the loss of high elevation specialists. In turn, this will have important consequences on alpine plant pollination.

Adaptively weighted maximum likelihood estimation of discrete distributions

SummaryDiscrete data arise in various research fields, typically when the observations are count data.I propose a robust and efficient parametric procedure for estimation of discrete distributions. The estimation is done in two phases. First, a very robust, but possibly inefficient, estimate of the model parameters is computed and used to indentify outliers. Then the outliers are either removed from the sample or given low weights, and a weighted maximum likelihood estimate (WML) is computed.The weights are determined via an adaptive process such that if the data follow the model, then asymptotically no observation is downweighted.I prove that the final estimator inherits the breakdown point of the initial one, and that its influence function at the model is the same as the influence function of the maximum likelihood estimator, which strongly suggests that it is asymptotically fully efficient.The initial estimator is a minimum disparity estimator (MDE). MDEs can be shown to have full asymptotic efficiency, and some MDEs have very high breakdown points and very low bias under contamination. Several initial estimators are considered, and the performances of the WMLs based on each of them are studied.It results that in a great variety of situations the WML substantially improves the initial estimator, both in terms of finite sample mean square error and in terms of bias under contamination. Besides, the performances of the WML are rather stable under a change of the MDE even if the MDEs have very different behaviors.Two examples of application of the WML to real data are considered. In both of them, the necessity for a robust estimator is clear: the maximum likelihood estimator is badly corrupted by the presence of a few outliers.This procedure is particularly natural in the discrete distribution setting, but could be extended to the continuous case, for which a possible procedure is sketched.RésuméLes données discrètes sont présentes dans différents domaines de recherche, en particulier lorsque les observations sont des comptages.Je propose une méthode paramétrique robuste et efficace pour l'estimation de distributions discrètes. L'estimation est faite en deux phases. Tout d'abord, un estimateur très robuste des paramètres du modèle est calculé, et utilisé pour la détection des données aberrantes (outliers). Cet estimateur n'est pas nécessairement efficace. Ensuite, soit les outliers sont retirés de l'échantillon, soit des faibles poids leur sont attribués, et un estimateur du maximum de vraisemblance pondéré (WML) est calculé.Les poids sont déterminés via un processus adaptif, tel qu'asymptotiquement, si les données suivent le modèle, aucune observation n'est dépondérée.Je prouve que le point de rupture de l'estimateur final est au moins aussi élevé que celui de l'estimateur initial, et que sa fonction d'influence au modèle est la même que celle du maximum de vraisemblance, ce qui suggère que cet estimateur est pleinement efficace asymptotiquement.L'estimateur initial est un estimateur de disparité minimale (MDE). Les MDE sont asymptotiquement pleinement efficaces, et certains d'entre eux ont un point de rupture très élevé et un très faible biais sous contamination. J'étudie les performances du WML basé sur différents MDEs.Le résultat est que dans une grande variété de situations le WML améliore largement les performances de l'estimateur initial, autant en terme du carré moyen de l'erreur que du biais sous contamination. De plus, les performances du WML restent assez stables lorsqu'on change l'estimateur initial, même si les différents MDEs ont des comportements très différents.Je considère deux exemples d'application du WML à des données réelles, où la nécessité d'un estimateur robuste est manifeste : l'estimateur du maximum de vraisemblance est fortement corrompu par la présence de quelques outliers.La méthode proposée est particulièrement naturelle dans le cadre des distributions discrètes, mais pourrait être étendue au cas continu.

Critical ruptures in a bundle of slowly relaxing fibers

We study the damage enhanced creep rupture of disordered materials by means of a fiber bundle model. Broken fibers undergo a slow stress relaxation modeled by a Maxwell element whose stress exponent m can vary in a broad range. Under global load sharing we show that due to the strength disorder of fibers, the lifetime ʧ of the bundle has sample-to-sample fluctuations characterized by a log-normal distribution independent of the type of disorder. We determine the Monkman-Grant relation of the model and establish a relation between the rupture life tʄ and the characteristic time tm of the intermediate creep regime of the bundle where the minimum strain rate is reached, making possible reliable estimates of ʧ from short term measurements. Approaching macroscopic failure, the deformation rate has a finite time power law singularity whose exponent is a decreasing function of m. On the microlevel the distribution of waiting times is found to have a power law behavior with m-dependent exponents different below and above the critical load of the bundle. Approaching the critical load from above, the cutoff value of the distributions has a power law divergence whose exponent coincides with the stress exponent of Maxwell elements

Experimental plant communities develop phylogenetically overdispersed abundance distributions during assembly

The importance of competition between similar species in driving community assembly is much debated. Recently, phylogenetic patterns in species composition have been investigated to help resolve this question: phylogenetic clustering is taken to imply environmental filtering, and phylogenetic overdispersion to indicate limiting similarity between species. We used experimental plant communities with random species compositions and initially even abundance distributions to examine the development of phylogenetic pattern in species abundance distributions. Where composition was held constant by weeding, abundance distributions became overdispersed through time, but only in communities that contained distantly related clades, some with several species (i.e., a mix of closely and distantly related species). Phylogenetic pattern in composition therefore constrained the development of overdispersed abundance distributions, and this might indicate limiting similarity between close relatives and facilitation/complementarity between distant relatives. Comparing the phylogenetic patterns in these communities with those expected from the monoculture abundances of the constituent species revealed that interspecific competition caused the phylogenetic patterns. Opening experimental communities to colonization by all species in the species pool led to convergence in phylogenetic diversity. At convergence, communities were composed of several distantly related but species-rich clades and had overdispersed abundance distributions. This suggests that limiting similarity processes determine which species dominate a community but not which species occur in a community. Crucially, as our study was carried out in experimental communities, we could rule out local evolutionary or dispersal explanations for the patterns and identify ecological processes as the driving force, underlining the advantages of studying these processes in experimental communities. Our results show that phylogenetic relations between species provide a good guide to understanding community structure and add a new perspective to the evidence that niche complementarity is critical in driving community assembly.